TORT IN A CONTRACTUAL MATRIX
John Fleming
Professor of Law, Emeritus
Boalt Hall School of Law, University of California
Our esteemed moderator sought to divide our discussion on the intersection between contract and tort, by allocating to me what he described as the
tort lawyer's view point. I found this puzzling at first but eventually
concluded that he sought to confine my comment on the way tort law has
been affected, whether by extending or contracting its reach, by a "contractual matrix",' by "the parties com[ing] together against a contractual
s t r u c t ~ r e . "My
~ role thus defined guards me against the possible charge of
harbouring a preference for tort solutions at the intersection between
contract and tort; indeed, far from this being the case, my preference for
the contract solution in, for example, overcoming the "privity gap" is
firmly based on the conviction that not only would it spare us collateral
problems (such as the applicability of limitation clauses), but that the
difficulty stems from a flaw of our contract law which can, and should, be
more expeditiously remedied on its own rather than by resort to tort.
But first, a preliminary observation. It is no exaggeration to say that the
tort law of the last twenty five years has been virtually obsessed with the
question of negligent economic loss. That exclusive focus tended to obscure that the defining element in many of these cases was the contractual
matrix in which the loss occurred. Latterly, however, the contractual
context has moved into the foreground of discussion and helped to elucidate
the real crux of the problem. That the situations most frequently involving
economic loss have their counterpart also in cases of property damage only
went to confirm that thesis. Thus in one of the construction cases, to be
presently discussed, concerning a claim by the owner against a subcontractor, the loss incurred was not, as in most such cases, the cost of repairing
the defect, but damage to adjacent p r ~ p e r t yAnother,
.~
somewhat similar,
recent case also involved damage to tangible property when a bailee's
servants caused structural damage to a warehoused tran~former.~
In both
cases the question was whether the prima facie duty of Donoghue v
Stevenson5 was negatived by the contractual matrix in which the relation
between plaintiff and defendant originated.
As an introduction to these " ~ o n t o r t " ~problems, the basic "concurrence" question should first be briefly addressed. English law, in contrast
to the French: has more recently taken the view that a contractual obligation between the parties does not preclude the concurrence of a tort duty in
the same respect, so as to give the injured party an option between pursuing
a contractual or tortious remedy. At first mainly encountered in relation to
claims for personal injury and property damage, application of this principle
1
2
3
4
5
6
7
I take this term from the judgment of La Forest J in London Drugs v Kuehne & Nagel [I9921 3
SCR 299.327.
Purchas LJ in Pacrfic Associates v Bawter [I9901 1 QB 993, 1023. Blom speaks of the "Interface
between Tort and Contract" in The Paisley Papers (1991) 139.
Nor~vichCity Council v Harvey [I9891 1 WLR 828 (CA).
London Drugs Ltd v Kuehne & Nagel [I9921 3 SCR 299.
[I9321 AC 562.
The term was coined by Gilmore, The Death of Coniract 90 (1974).
See Weir, Internaiional Encyclopedia of Comparative Law XI12 ch 12. .
270
Canterbury Law Review [Vol. 5 , 19931
has been greatly increased by the contemporaneous expansion of negligence liability for economic loss, especially in its application to profes. ~ motivation
sional services by accountants, bankers and l a ~ y e r s The
behind this development seems to have been to make available to the
claimant procedural and other advantages of tort, such as contribution or a
more favourable starting point for the period of limitation, rather than some
But such tort duty is subject to any
more profound doctrinal convi~tion.~
contractual modification, like a lower standard of care or limitation of
liability, in deference to the belief in the primacy of private ordering.'' In
short, the starting point for tort in a contractual matrix is that "contract
trumps tort" (to borrow Peter Cane's pithy axiom)."
The attraction of tort remedies has occasionally lured courts into more
extreme ventures, as in the one-time suggestion in California of treating all
malicious breaches of contract as tortious, so as entitle the aggrieved party
to general damages for mental distress or even punitive damages.I2 But this
"tortification" was later confined to bad-faith refusals by insurers to settle
claims against their clients.I3 So limited, it has even persuaded aNew South
Wales judge that it was proof against summary di~missal.'~
Much more complex is the "triangular" situation: the effect, if any, of
a contract upon the possible tort remedy of or against an outsider. None
less than Donoghue v Stevenson seems to teach that the days have gone
when a contract duty to A precluded a tort duty to B in the same matter.
The manufacturer's contractual promise to his buyer regarding the quality
of the product no longer pre-empts an obligation to the ultimate consumer
in tort. It is the universality of this principle which was being questioned
in the two above-mentioned cases, to which we must now return.
In the construction cases the building owner contracts with a head
contractor for the work to be performed. A subcontractor causes damage
or other loss to the structure, for which the owner is seeking recovery from
him. Paradigmatically, the subcontractor is under contract with the head
contractor, but not with the owner who suffered the loss. The reason why
the owner does not address his complaint to his contractual partner, the
head contractor, may be the latter's impecuniosity (or bankruptcy), a
common feature in a lightweight industry. Most of the recent cases have
categorically denied any duty of care between these parties on the ground
that the loss is purely economic when the claim relates to a deficiency in
8
9
lo
II
12
13
14
Midland Bank v Hett, Stubbs & Kemp [1979] Ch 384 (solicitor); Esso Petroleum Co v Mardon
[I9761 Q B 801 (landlord); Mouat v ClarkBoyce [I9921 2NZLR 559; Central TrustvRafuse [I9861
2 SCR 147. In England, the question of concurrence is now arguable only in the House of Lords:
Nitrigin Eireann Teoranta v Inco Alloys [1992] 1 All ER 854, 859;'
Mustill LJ recently described differing periods of limitation as offend[ing] common sense":
SociPtP Commerciale de Riassurance v ERAS [I9921 2 All ER 82, 85. Compare Lord Scarman's
discouraging statement in Tai HingCotton MiNv Liu ChongHingBank [l986] AC 80, 107: "Their
Lordships do not believe that there is anything to the advantage ofthe law's development in search
for aliability in tortwhere the parties are in acontractual relationship." It occurred inacase denying
a tort duty of care after the same duty had been rejected as an implied term of the contract. See
generally Cane, Tort Law and Economic Interests 336-344 (1991).
See Central & Eastern Trust Co v Rafuse [I9861 2 SCR 147; British Columbia Iiydro and Power
Authority v BG Checo International Ltd 119931
- 1 SCR; 99 DLR (4th) 577.
Cane, op cit.
Seaman's Direct Buying Service v Standard Oil Co (1984) 36 Cal3d 752.
Foley v Interactive Data Corp, 47 Cal2d 654 (1988).
Gibson v The Parkes District Hospital (1991) 26 NSWLR 9. See my comment in (1992) 108 LQR
357; also Handford, (1993) 1 Tort L Rev 87.
Tort in a Contractual Matrix
27 1
the work done.15 That generalization is foreign to my topic except to the
extent that it may rest, in this context, on the residuary argument (to be
discussed in the second part of this paper) that the plaintiff could have
planned in advance to shift the risk by contract. Here I am concerned more
specifically with cases where the contract between the building owner and
the head contractor,16or between the head and subcontractor,17purported
to exclude or limit liability. This situation is by no means limited to claims
for purely economic loss, but can as well arise, as one case has already
shown,18in relation to property damage -where, in other words, "duty"
is prima facie governed by Donoghue v Stevenson rather than by any
overriding constraints against recovery for negligent economic loss.
Can the building owner sidestep such a limitation clause by suing the
subcontractor in tort? It would be unrealistic to reject such a claim outright
on the ground that there is no "proximity", the currently fashionable
touchstone of "duty", between the parties. The employment of subcontractors was clearly contemplated, and the risk of their negligence and its
effect on the owner's interests was obvious. But was it "just and reasonable" (to use another favoured test of duty) to subject the subcontractor to
liability in these circumstances, which included the contractual background
against which he had agreed to perform his services? Negating a tort duty
can be squared with conventional analysis as an illustration ofthe principle
of voluntary assumption of risk: in as much as the owner had agreed with
the head contractor to assume the risk of defective performance above the
stipulated sum, he assumed the risk of loss from negligence by the contractor and those whom the latter deputed to perform the work. If, instead, the
limitation clause was in the contract between the subcontractor and the head
contractor, it signified all the more the subcontractor's unwillingness to do
the job otherwise than subject to the limitation. At any rate, if known to the
owner, the latter's acquiescence can be deemed an acceptance of the terms
under which the subcontractor is alone prepared to enter into a relationship
defining his duty with the owner.
But could a tort duty be recognised subject to the limitation clause in the
contract? This baffled the judges. In The Aliakmon,lg where the consignee
of cargo sued the shipowner who had contracted with the consignor subject
to the Hague-Visby Rules, Donaldson M R gave it as his principal reason
for denying a tort duty, and Oliver L J as an alternative reason, that a tort
duty would be "original" (not derivative) and as such could not be subject
to the limitation clause in the bill of lading. Only Robert Goff L J was
prepared to invoke the concept of "transferred loss" and thereby recognise
~
as no surprise
a tort claim subject to the contractual l i m i t a t i ~ nIt. ~comes
that this imaginative solution, following the German model of
Dritt~chadensliquidation,~~
was promptly squashed by Lord Brandon in the
15
16
17
18
19
20
21
Based on Murphy v BrentwoodDC 119911AC 398, categorising such loss as economic. Such cases
include Simaan General Contracting v Pilkington Glass Co [I9881 QB 758 (CA) and Greater
Nottingham Co-op v CementationPiling and Foundations [I9891 QB 993 (CA). Exceptional was
Junior Books v Veitchi Co 119831AC 520 which is now deemed aberrational.
Southern Water Authority v Carey [I9551 2 All ER 1077 (CA).
Norwich City Cotrncil v Ifarvey [I9891 1 WLR 828 (CA).
Norwich City Council v Harvey, supra.
[I9851 QB 350.
The sane result had earlier been contemplated by Lord Roskill in the controverted case of Junior
Books v Veitchi Co [1983] 1 AC 520,546. The precise problem in The Aliakmon, the position of a
buyer whose seller retained title at the time of the damage, has since, with Lord Goff s blessing,
been removed by giving him a right to sue under the Bill of Lading Act 1992. See Bradgate &
White, (1993) 56 Mod L Rev.188.
See Markesinis, "An Expand~ngTort law - The Price of A Rigid Contract Law", (1987) 103 LQR
354,387 ff (1987).
Canterbury Law Review [Vol.5 , 19931
House of Lords.22It does, however, still enjoy apparent support in Canada.23Whichever alternative, its significance is linked to the view that the
operation of tort law in the context of a planned transaction calls for a
modification of the rules familiar in the "classic" non-consensual situations
like the typical traffic accident.
Less problematical, no doubt, would be a contractual solution. This is
preferred by American courts, either on a theory of assignment or third-party
beneficiaries, for claims against subcontractors and the like.24 It neatly
disposes of the problem of limitation clauses by respecting the contractual
allocation of benefits and burdens. At present, however, this solution is
blocked for British courts by the rigidity of the privity doctrine. Its adoption
is strongly recommended in the academic l i t e r a t ~ r e . ~ ~
The other, previously mentioned illustration of a tort duty modified by
a contractual matrix comes from the important pronouncement of the
Supreme Court of Canada in London Drugs v Kuehne & N ~ g e l The
.~~
plaintiff had delivered a transformer to a warehouse company for storage
under a standard form contract which stipulated a limitation of liability on
any package to $40. Two employees while trying to move the transformer
with forklift trucks caused it to fall, resulting in severe damage. Could the
plaintiff circumvent the limitation clause by suing the employees in tort?
Unlike the preceding cases against subcontractors who, presumably,
were at least covered by liability insurance, to permit a claim against
employees would seem at first glance particularly unfair, if not indeed
incongruous with modern understanding of labour relations. The cost of
accidents, it is generally believed, is a charge on the employer, since he,
unlike the employee, is able to plan against, and absorb, it as an overhead
of his operations. Nor is it surprising that so few cases have actually raised
the tort liability of employees because they are rarely worth suing. It would
only be in exceptional circumstances that employees have insurance of
their own (eg on their own car) or otherwise command sufficient financial
resources to answer a judgment. Thus the argument that it would be unfair
to deny the injured plaintiff redress against the guilty employee misses the
target, since he could not count on the employee's ability to answer for the
damage. The rare case, like presumably the present, where the employer's
insurance policy also covered his employees, does not justify a contrary
conclusion.
22
23
24
25
26
[I9861 AC 785.
BC Hydro & Power Authority v ND Lea &Associates (1992) 92 DLR (4th) 403,438-9, noting the
apparent consensus of the Court of A ~ o e a in
l London Drum (1990) 70 DR (4th) 5 1 that contractual
t e h s may limit the tort duties of
who are not
to thecontract: t his view was restated
by McLachlin J in the Supreme Court (see infra p 273). It is beyond doubt that contractual terms
bind parties in cases of concurrent contract/tort liability (supra at n 11).
See William K Jones, "Economic Losses Caused by Construction Deficiencies: The Competing
Regimes of Contract and Tort" 59 Cin L Rev 1051, 1077-1 101 (1991); Eisenberg, "Third-Party
Beneficiaries", 92 Col L Rev 1358, 1 4 0 2 6 (1993). The tort solution is generally, though not
uniformly, rejected on the ground that the loss is economic.
The Supreme Court of Canada affirmed an 'obligation pour autrui' in Quebec law: Demers v
Dtrfresne Engineering [1979] SCR 146, though that is more doubtful in French law: Wallace,
Consbuction Contracts 5-22 to 5-26 (1986). Markesinis, The German Law of Torts (2d edn 1990)
adds that "it has never been doubted in German law that the contractual debtor can oppose against
the third partylplaintiff all defences etc he may have against the contractual creditor."
Markesinis, "Eternal and Troublesome Triangles, 106 LQR556 (1990); Beyleveld & Brownsword,
Privity, Transitivity and Rationality" 54 Mod L Rev 48 (1991). The French theory of "groups de
contrats", in effect advocated by the latter, has however since been dismantled by the Cour de
Cass. (Ass plen) 12 July 1991 (Rec Dalloz D) 1991, 549; 1993 ZEuP 592.
supra, n 1.
Tort in a ContractuaI Matrix
273
Likewise, from the employer's point of view -he cannot be expected
to bail out the employee in the circumstances. His contract with the owner
of the transformer, limiting his liability to $40, meant that it was up to the
latter, not himself, to procure excess insurance. It would thus upset the
whole pattern of the warehouse agreement for the employer, after all, to be
saddled with the cost of the accident.
Nor, finally, would it be defeating the plaintiffs expectations to be
limited to $40. Obviously, the warehouse contract envisaged the participation of employees, and it would be quite unrealistic to suppose that the
plaintiff, while assuming the risk of loss in excess of $40 vis-a-vis the
warehouseman, reserved recourse against the latter's employees. Indeed,
the benefit of such a manoeuvre, were it to succeed, would enure not for
him, but for his insurer, who would be subrogated to his claim against the
hapless employee. Thus, once again, the intendment of the whole planned
arrangement would be frustrated, not even for the benefit of the tort victim
but for his insurer who would gain a wi~idfallby being enabled to shift to
another party a loss which he had contracted to underwrite. From every
functional point of view, then, the claim against the employees should fail,
at least in excess of $40, if not altogether.
The majority adopted the contract approach
And so it did. But
by facing down the privity rule and allowing a limited jus tertii to the
employees. They should be entitled to invoke a limitation clause in the
contract between their employer and the plaintiff if (1) that clause expressly
or impliedly extends its benefits to employees seeking to rely on it; and (2)
the employees acted in the course of their employment and in the performance of the very services provided for in the contract when the loss
occurred. The remaining three justices preferred the tort approach. One
justice (McLachlin J) expressed her unease, not so much with modification
ofthe privity doctrine as with the finding of the majority that the employees
were implied beneficiaries of the limitation clause, seeing that they had not
been so much as mentioned in the contract. To give them the benefit of
such a clause was to travel far beyond the so-called Himalaya clauses,
which had eventually gained recognition by the Privy Council,28and could
most probably not have been accomplished even under statutes (such as
New Zealand's Co~itracts(Privity) Act 1982) giving the benefit of contract
promises to "intended beneficiaries". She preferred the "duty" approach
from the above mentioned English cases. Essentially based on voluntary
assumption ofrisk, it recognised, however, that aduty ofcare can be waived
or qualified in the light of the contractual matrix. By agreeing to the
limitation clause, the customer had accepted that the risk of damage in the
performance of the contract was to be his, and his alone.
The most drastic departure from orthodoxy was La Forest J's challenge
of the putative liability of employees for torts committed in performing
their employer's contractual obligations. In cases like the present, involving a planned transaction with the employer, the plaintiff can be considered
to have chosen to deal with a company. The employees were not given to
understand that the plaintiff was relying on them for compensation. Placing
liability exclusively on the employer places liability on a party that is easily
27
28
For more extensive treatment see my Comment on London. Drugs in 13 OJLS 430 (1993); also
Huscroft & Manning, 1993 N Z Recent Law 210 (1993).
NZShipping Co v Satterth~z~arte
(The Eutymedon) (19751 ACI 54; Port Jackson Stevedoring Co v
Salmond [I9811 1 WLR 138; 144 CLR 300; ITO-International Terminal Operators v Miida
Electronics [I9861 1 SCR 752.
274
Canterbury Law Review [Vol.5 , 19931
able to modify its exposure by contractual stipulation. In contrast, its
employees had no real opportunity to decline the risk, nor could their union
be expected to bargain over an issue that so rarely assumes practical
significance. La Forest J sympathized with the position of several legal
systems, including statutes in New South Wales, South Australia and the
Northern Territory, which have totally abrogated the tort liability of employees, but he prudently confined his holding, within the parameters of
the case before him, to what he called "contractual vicarious liability", ie
cases within a contractual matrix.
La Forest J's reasoning led him into dissent from the conclusion of the
rest of the Court that the defendants were liable but only for $40 per
package. His view of the employees' tort immunity, based on policy rather
than derived from the contract itself, meant that the employees were not
liable at all. In other words, they were not pro hac vice identified with their
employer any more than with his contract. This conclusion more nearly
reflects the commonly held impression, reinforced by the public policy
behind vicarious liability, that employees should not incur any liability at
all for the torts committed by them in the course of employment but be
subject merely to disciplinary sanction by their employer. On that view,
the contractual setting should be irrelevant; in which case, neither the
majority's preference for ajus quaesitum tertio nor La Forest J's limitation
of employees' tort immunity to a contractual context looks the ideal
long-term solution; the latter proves acceptable only as a step in the right
direction.
I come now to a second situation relevant to defining the scope of tort
in the light of contract.
A powerful argument, heard of late, purports to deny a duty of care,
particularly a duty to avoid pure economic loss, if the claimant should have
exercised his ability to guard against the risk by self-protective measures
such as by contract or insurance. It was first voiced in The A l i a k m ~ nin
,~~
support of the decision by the House of Lords (it will be recalled) to deny
a tort remedy against a negligent carrier to a consignee to whom the risk
but not the property in the goods had passed at the relevant time. Lord
Brandon offered the buyers the gratuitous advice that they should have
contracted with the sellers to exercise the right to sue the carrier for their
account or assign such right to them.30
Lord Brandon's suggestion has struck a responsive chord in later cases.
It has, for example, played a role in the construction area, in denying tort
recovery to a building owner against a subcontractor or to a contractor
against the owner's supervising engineel3' on the reasoning that the claimant could have contracted with the defendant for added protection; indeed
his failure to do so implied an unwillingness to enter into formal relations
29
30
31
Leigh & Sillavan Ltdv Aliakmon Shipping Co [1986] AC 785,819. See Cane, op cit at 345-8, with
citation of literature.
But this advice stopped short of recognising an equitable duiy to sue on behalf of the consignee,
perhaps in deference to an obiter remark by Lord Diplock in The Albazero [I9771 AC 774, 845.
Arguing in favour of such a duty is Goode, "Ownership and Obligation in Commercial
Transactions" 103 LQR 433,453 ff.
On the measure of damages in asuit by the consignor see GUSProperiy Management v Littlewoods,
[I9821 SLT 533 (HL); Linden Gardens Trust v Lenesta Lid [I9931 3 WLR 408 (HL).
Simaan v Pilkington [I9881 QB 758 (CA); Pac$c Associates v Buxter [I9901 1 QB 993 (CA).
Tort in a Contractual Matrix
275
with 11im.~~So,
accountants have been absolved from liability to investors
and even to existing shareholders on the ground, among others, that such
plaintiffs should have procured independent advice.33The argument is also
encountered in its negative form, namely that the plaintiffs inability to plan
against the contingency of loss favours recognition of a duty of care by the
defendant. Hence the solicitor's liability in tort to the testator's intended
beneficiary, where the latter's inability to guard against the loss reinforced
the desirability of furnishing a sanction against professional n e g l i g e n ~ e . ~ ~
Also in the remarkable decision of Smith v
where a mortgagee's
valuer was exceptionally held liable to the purchaser of a "modest" home
for an overvaluation, in view of her unfamiliarity with business matters
which excused her failure to contract for independent advice such as one
would have expected from a business person.
Yet despite its evident appeal, the argument proves too much. For one
thing, it does not explain why it was not invoked in Hedley Byrne v HellePj
or at least subsequently added as a defining element of liability for negligent
mi~representation.~~
Where it did surface, as in C a p a r ~ ;it~ was just a
make-weight to reinforce the principal ground, which was that the plaintiffs
constituted too undefined a class to satisfy the element of "proximity".
a true instance of a tort claim in
Nor was it confronted in the Norsk
a contractual matrix. The defendants, negligently navigating a barge,
collided in heavy fog with a bridge owned and operated by a public
authority. The plaintiffs, under license from that authority, were the principal users of the bridge, who sued for their loss caused by the disruption
of its use. A majority of the Supreme Court of Canada, casting aside prior
English authority, allowed the tort claim against the bargees, notwithstanding the availability of contractual planning against that risk with their
contractual partner, the owner of the bridge. Yet, as the dissent pointed out,
the plaintiffs were in a much better position to assess that risk than were
the defendants, and could easily have shifted it to the bridge owner.
Moreover, the postulate is difficult to reconcile with there being no rule,
certainly no general rule, that even the existence of an alternative cause of
action militates against, let alone precludes, recognition of a tort duty by a
remoter party.40The "ultimate consumer" does not forfeit his tort claim
against a negligent manufacturer by having a cause of action for breach of
warranty against the retailer. It is precisely because such an alternative
remedy is ineffective (eg because of bankruptcy) or to avoid circuity of
actions that a plaintiff seeks recovery from the remoter tort defendant, and
~ '
it is arguable that "a linear arrangement of
is allowed to do ~ 0 . While
A fortiori, if there was a contract between plaintiff and defendant, which however dealt with
extraneous matters, as in Greater Nottingham Cooperative Society v Cementation Pilings and
Foundations [I9891 QR 71.
Caparo Industriesplc v Dickman [I9901 2 AC 605.
White v Stone [I9931 NLJ 473 (CA); Ross v Caunters [I9801 Ch 297. See my note, (1993) 109
LOR 344.
[1990] 1 AC 83 1.
[I9641 AC 465. Cf Stapleton, "The Duty ofcare and Economic 1,oss: A Widcr Agenda" 107 LQR
249, 259 (1991): "So, far from correcting a 'wrong turning', Ifedley Byrne judged by today's
standards may have been one itself."
Or does it lurk in the requirement that reliance be reasonable?
supra, n 33.
Norsk Paclfic Steamship Co v (7anudian National Rail~uay[I9921 1 SCR 1021. See my comment
at (1993) 1 Tort L Rev 68. Nor, it might be added for the Australian reader, in Caltex 011v The
Dredge "Willemstud" (1976) 136 CLR 529.
Nor docs a contractual remedy preclude an alternative tort cause of action for breach of the same
obligation against the same defendant: supra, at n 8.
If such a remedy against a contractual partner is actually available and would be the appropriate
Canterbury Law Review [Vol.5, 19931
obligations is preferable in order, for example, to shield peripheral deeppocketed parties and/or to control the problem of multiple c a ~ s a t i o n , " ~ ~
this is an agenda to which many concerned with plaintiffs' rights would
take staunch objection.
Perplexing also is how to square the "Brandon argument" with the
justification for recognising a tort duty on the ground that the relation
between the parties was "akin to contract". For if that was the case, why
didn't they "go the extra mile" and contract?
In Hedley Byrne, for example, the relationship between the defendant
bank and the plaintiff to whom the former had furnished, through his own
bank's mediation, a credit report, was described as "the equivalent to
contract".43 In the controversial decision of Junior B o o b v Veitchi C O , ~ ~
allowing an owner to sue a subcontractor in tort for applying defective floor
finish which subsequently needed replacement, the relation between the
parties was described as "only just short of direct contractual relationship.'' And though that decision has been widely disparaged in later
where the purchaser of the
the argument was heard again in Smith v
modest home was deemed to be in a relation "akin to contract" with the
valuer employed by the prospective mortgagee. Admittedly, the fact that
the cost of the valuation was debited to her account confirmed the reality
of that description. In none of these cases was it so much as hinted that the
plaintiff could have availed himself of the opportunity to protect himself
by contract.
Again, why has the "Brandon argument" been confined in the main to
claims for economic loss? Is it not just a cover-up for a peremptory policy
against negligent economic loss? It is no doubt true that in most cases of
personal injury the victim is in no position to protect himself in advance
by contract, except in the unacceptable sense of procuring general accident
assurance. Self-protection is germane principally to the context of planned
transactions, which rarely involve personal injury. It can, however, arise in
relation to property damage, as illustrated by the following variant of the
"Himalaya clause" cases.47A carrier contracting with the consignor of
goods to be shipped by sea stipulates in the bill of lading for limited liability.
The cargo is damaged due to the negligence of a stevedore. Can the
consignee sue the stevedore? The stevedore's duty in tort certainly arose
42
43
44
45
46
one to deal with the matter, South Paclfic Manufacturing Co v NZSecuriV Consultants [I9911 2
NZLR 282 suggests this as a factor militating against a tort duty by a third party. The CA disallowed
a tort action against an investigator, employed by the plaintiffs insurance company, who had
incriminated the plaintiffof arson. Only Richardson J went so far as to say that "ifthe insured have
a remedy in contract against the insurer they should exercise that remedy. If they do not have an
adequate remedy that is because they only paid a premium which gave them that lesser protection.
In that situation I cannot see any justification for allowing them a greater recovery through tort
than they were prepared to pay for in contract" (at 308). In another case, Gran Gelato v RichclifSe
Group [I9921 1 All ER 865 the Vice-Chancellor went even further in denying a remedy against a
seller's solicitor who had made a direct misrepresentation to the purchaser on the ground that the
later had a contractual remedy against the vendor (who however was bankrupt). The decision has
been castigated by Cane, 109 LQR 539 (1992); and I-lirst J in The Nicholas H [I9931 2 Lloyd's
Rep 481, 497 considered that the Vice-Chancellor had not intended to "lay down any general
principle."
Stapleton, "The Condition of the Law of Tort'' (a yet unpublished lecture delivered at All Souls
College, Oxford on 5 July 1993) 33. The author advocates this among several other countervailing
factors against the imposition of tort liability.
Hedley Byrne v Heller [I9641 AC 465, 530.
119831 1 AC 520.533.
billon LJ went so far as to say that "I find it difficult to see that future citation of the Junior Books
case can ever serve any useful purpose" (Simaan v Pilkington [I9881 Q B 758,784).
119901 1 AC 831. 796.
Tort in a Contractual Matrix
277
in a contractual matrix and arguably should be modified by the consignee's
ability to contract against the risk of loss. This supposition is strengthened
if the carrier in the bill of lading actually stipulated the exemption for
himself, his servants and agents. Regardless of whether the benefit of such
a clause would bind the consignee in contract, should he not be defeated
by the argument that he actually exercised his opportunity of dealing with
the risk by waiving his tort rights against the negligent stevedore? This
formulation returns us to the "assumption of risk" explanation for qualifying a duty of care in the light of an exemption clause to which the plaintiff
was not privy.48 Or, in Jane Stapleton's words, "a duty would not be
allowed in circumvention of a positive arrangement regarding the allocation of risks which had been accepted by the la in tiff."^^
Another difficulty with the "Brandon argument", as Peter Cane points
out,50is the lack of definition in the supposition that the plaintiff was able
to protect himself by contract. Is that formula not just as unstable as the
once-touted factor of reliance?jl This was a central issue in the House of
Lords decision of Smith v Bush, previously mentioned,j2 in which a valuer,
acting on behalf of a mortgagee, was held liable to a purchaser who had
been shown and relied on the overvaluation. Special stress was laid on the
transaction involving a "modest dwelling" and the fact that the woman
could not be credited with familiarity with business affairs. While this
approach is entirely praiseworthy for its sensitivity to social realities and corresponds with the distinction drawn by legislation in England between
the rights of subsequent buyers of homes and commercial property against
the original builder53 - it is difficult to reconcile with Lord Brandon's
insistence on observing a "clear bright line" in cases of economic loss.54
Still less is there any justification for requiring the plaintiff to protect
himself by contract, when the defendant had an equal opportunity of
contracting for exemption. This was the situation in Edgeworth Construction
v Lea & Associates5j where a building contractor relied on faulty specifications supplied by the owner's design engineers. The latter could have
placed a disclaimer on the documents, which Hedley Byrne teaches would
have been effective vis-a-vis the contractor. There was even less reason,
therefore, for requiring the latter to protect himself by contracting for
protection with his own employer, the building owner. Nor did the fact that
that contract disclaimed the employer's responsibility for the engineering
specifications oust the engineer's duty because the contract -the contractual matrix - did not purport to define the engineer's obligations.
Jane Stapleton proposes to elevate the inadequacy of alternative means
of protection to a necessary, but not sufficient pre-condition to liability for
economic loss.56 In view of the preceding list of reservations, such a
generalisation strikes me as precarious. The "Brandon argument'' makes
its strongest appeal to those who believe in the primacy of private ordering,
expressing the ideology of the free marketeers of the 80s. The law of torts,
supra, at p 27 1.
"Duty of Care and Economic Loss: A Wider Agenda" (1991) 107 LQR 249, at 296
so OD cit at 350-1.
51 see Stapleton, supra, n 49 at 2 8 3 4 .
52 supra, n 35.
53 The Defective Premises Act 1972 creates a warranty for successive owners of dwellings, though
subject to ashort period of limitation. Only commercial buyers are affected by Murphyv Brentwood
DC [I9911 AC 398.
54 The Aliakmon, supra at 817, 820.
5s [I9931 3 SCR 206.
56 supra, n 49 at 295.
48
49
278
Canterbury Law Review [Vol. 5, 19931
on the other hand, represents historically and functionally a counterstrain
of public policy which on occasion overrides private agreement. A striking
example is, of course, the voiding of exclusion or limitation clauses on
grounds of public policy. Donoghue v Stevenson itself demonstrates how
permeable and shifting is the frontier between contract and tort, culminating in the triumph of social policy (tort) over private ordering (contract).
Despite occasional, temporary retreats (such as we have been witnessing
in recent years in England), that has been the momentum of change in
modern law, not the direction signalled by Lord Brandon.
Stapleton's focus was on economic loss, attempting to narrow down to
four the factors relevant to a delineation of negligence liability for economic loss. But whatever one's reservations to the generality of her thesis
("necessary though not sufficient") in that wider context, a much stronger
case can be made for the importance of that factor in the setting of planned
transactions, which she did not address. This for the obvious reason that in
such a contractual matrix significance necessarily attaches to a party's
opportunity to consider his own position with regard to the risks confronting him. Such were the building construction cases, like Simaan v Pilkingtod7 and particularly Pacijic Associates v B a ~ t e rIn
. ~the
~ latter case,
the plaintiffs had contracted with their employer to carry out works to be
supervised by the defendant engineers. That contract contained a disclaimer of liability by the defendants to the plaintiffs. In an action by the
contractors against the engineers for their refusal to certify claims, the
Court of Appeal ruled against any duty of care. The disclaimer clause
showed that the defendants expressly declined to assume responsibility to
the plaintiffs. Thus far from having availed themselves of the opportunity
to shift the risk by contracting with either the employer or the engineer, the
plaintiffs had actually expressly taken the risk unto themselves. Here then
the role of contract, whether the plaintiffs or the defendant's with a third
party, assumes a critical function in the formulation of "duty" to which
the plaintiff is entitled for protection.
The history of tort in the matrix of contract stretches back to Winterbottom v Wright,59 if not beyond. For a time it became accepted that a
defendant's contractual undertaking pre-empted a corresponding tort duty
not only to the contractual partner but also to third parties. Donoghue v
Stevenson60 exploded this "privity fallacy" in relation to claims for personal injury and damage to property. That the defendant owed a contractual
obligation to A did not preclude his owing a tort duty in the same matter
to B. But other constellations of the contract matrix raised new problems,
mainly in the context of purely economic loss, once the barrier to such
~ 'common feature
claims was slightly set ajar by Hedley Byrne v H e l l e ~ .A
is their setting in a planned transaction.
One such problem concerns the effect of a contractual limitation clause
on the tort liability of a third party. This can be resolved either by expanding
the reach of contract beyond privity or by "jumping the privity gap" in
57
suprqn31.
ss supra, n 31.
59
60
61
(1842) 1 0 M & W 109; 152ER402.
supra, n 5.
supra, n 36.
Tort in a Contractztal Matrix
279
tort. The first part of this paper purported to deal with various attempts to
follow the tort route. Whether this is preferable to the contract approach is
quite another matter. It might be interesting to speculate why British courts
adopted the one, while American and Canadian courts inclined to the other.
The answer, I believe, lies in the comparative flexibility of their contract
and tort rules, as respectively perceived. Whatever might be said in general
about the adaptability of the modern law of contract to changing social
demands,62the privity rule has, in British perspective, assumed a unique
aura of solidity (stolidity) despite its relatively modern provenance. By
contrast, tort law has become free-flowing despite occasional cyclical
setbacks such as it is currently experiencing. Resort to tort is therefore seen
as the only feasible avenue to the desired outcome, in much the same way
as conversely German law has been forced into contract routines in order
to escape the straitjacket of its outdated Code regime of tort. By contrast,
American and Canadian courts have felt freer to choose between contract
and tort without comparable doctrinal constraints. In the present context,
their preference for the contract approach is undoubtedly due to the
perception that it is the more germane and effective way of dealing with a
matter that arises in a matrix of contract and does not call for a last resort
remedy of tort.
The other problem in the context of a contract network we have considered is what effect to attribute to a plaintiffs failure to avail himself of an
opportunity to protect himself in advance by contracting with the injurer
or a third party. This raises an issue of a different order, not the more or
less technical one of which road to travel to an agreed destination, but one
of substantive policy whether to deny the protection of tort law to a person
capable of advance planning against the risk. The emphasis on individual
responsibility in the political and economic ideology of the recent past has
given a lift to this argument, but its place in received doctrine must needs
await further testing.
Both situations illustrate the weakening of the borders between contract
and tort which were once so clearly delineated by the classical theorists of
both disci lines. Twenty years ago Grant Gilmore presaged the Death of
Contract,i' noting the on-going absorption of contract into tort as the result,
among others, of the socialisation of contract law, once the stronghold of
private ordering free from intrusion by public policy. But that was, or is,
only part of the story. In reality, the interaction has become reciprocal; if
contract is being obliterated in some contexts, like concurrence, in others
it has come to define the incidence of tort duties. This is especially so in
the context of planned transactions, which until yesteryear fell within the
exclusive domain of contract. As tort has expanded its reach into that
sphere, it learnt to respect the important message of private ordering.
These developments surfaced at a time of growing judicial concern for
ways to rein-in the expansive, indeed explosive, potential of tort liability,
especially of the law of negligence. Indeed, my friend Jane Stapleton
recently sounded a call for a change of approach: from the Anns strategy
of, in effect, asking "Why not?" to focusing on the "countervailing factors
which weigh against the assistance of tort."64 This paper is a contribution
to that debate.
'
62
63
A generally optimistic view was recently taken by ~ c ~ a c h l a "The
n , 'New' Law of Contract in
New Zealand" [1992] NZ Recent L Rev 436.
(1974).